1 Introduction (~6-7)1.
1 Motivation ()1.2 Technologies surveyed ()1.2.1 Network Slicing1.2.2 NFV1.2.3 SDN 1.
3 Taxonomy ()2 Analysis of existing solutions (~10)2.1 limitations2.2 Key issues 2.
3 Alternatives on how key issues can be solved with the advancements 3 Conclusion (~3)IntroductionThe goal of this paper is to introduce the reader to the advancements of the fifth generation of mobile networks (5G) and its core architecture in terms of research but also implementation, by surveying the ongoing standardization process and the in parallel development of first proof of concept implementations by the mayor industry leading mobile service providers. The outcome of this paper will therefore be a review of the current status in development and deployment of those mobile networks.MotivationFor the last years, the development of the new fifth generation of mobile networks has been a driving factor in research as well as the industry. The fifth generation of mobile networks has to be dynamic to cope with the huge demand of the users to increase the capacity, provide higher data rates, decrease latency, as well as providing enhanced quality of service.
All of that is the condition to be applicable in the various different use cases such as the massive machine type communications with the Internet of Things and billions of devices being connected to the mobile networks. Also, the enhanced mobile broadband that has to support all the services like video streaming in ultra-high definition or virtual reality applications, by providing ultra-high throughput capabilities. The third use case domain identified is the ultra-reliable and minimum latency type of communications especially important in mission critical applications and emergency situations, with latencies less than 1ms.Technologies surveyedIn the new world of 5G new technologies are needed to enable the flexibility and customizability needed by all the different applications and services that have to be supported.
While the technology of current 4G networks is mostly specialized equipment, which is designed to be optimized for a specific application, 5G networks have to adapt to totally different requirements in the different use cases. Therefore, it is crucial to transform the specialized hardware components into flexible and customize software components. These software components then have to be composed together so that they together form a so-called network slice to ensure the best performance for a certain domain of services or applications that share a similar set of requirements. This ability is called network slicing.
Technologies like cloud computing, software defined networking and network function virtualization are the key enabling technologies to realize the network slicing. Cloud Computing, which is the virtualization of hardware resources is the key technology to provide a virtualized infrastructure. Network function virtualization enables the virtualization of network functions that formerly run on specialized hardware and will now be operated in a cloud environment.
The most popular architecture design is the ETSI Reference Architecture for NFV(ETSI GS NFV 002 v1.2.1). The third enabling technology is software defined networking (SDN). SDN is based on the idea of decoupling the data plane from the control plane, which gives the ability of flexibly controlling the network forwarding entities to the network service provider. This architectural concept is also adopted in the standardization and development of the 5G network core architecture.
NFVDefinitionNetwork Function Virtualization (NFV) is a network architecture concept that enables the use of technologies of IT virtualization to virtualize entire parts of network endpoints into blocks that can be stick together into virtual networks.12 It’s meant to minimize costs for network operators by dissociate functions or services like a Load balancing, DNS and Security services from physical servers to a virtual ones.13 GoalNFV aims to change the way of network operator’s design by improving the standard IT virtualization technology concept.12 It involves the implementation of network functions in software that can run on a range of industry standard server hardware, and that can be moved to, or instantiated in, various locations in the network as required, without the need for installation of new equipment. 1015Advantages and BenefitsNFV reduces the hardware demand required for the establishing networks by transfer network functions and services into software to be editable virtually from anywhere within the operator’s network. As result, plentiful of benefits for the network operator arise such as, Reducing equipment costs, with flexible network function definition and allocation is an important enabler for network adaptability.
Application and Use CasesNFV can be deployed to any data packets processing and control plane function in different types of networks. Some application and use cases for NFV can be such as: Mobile Network nodes, Traffic Analysis, Virtual Networks, Data Center Micro Segmentation and Network Functions as a Service.ChallengesImplementing NFV is interesting but may face some challenges because of lack of standards in NFV management, automation and orchestration (MANO). MANO is a framework for managing virtualized network functions (VNF). The European Telecommunications Standards Institute (ETSI) Industry Specification Group (ISG NFV) defined the MANO architecture to facilitate the deployment of services as dissociate physical devices to Virtual machines.
Funding the new technology will be also another challenge that operators face while implementing NFV.15 Software-defined networking (SDN)DefinitionSoftware-defined networking (SDN) has been defined in several ways, originally defined an architecture to design and manage networks that separate each of network control and data forwarding, Where a control level can run several devices. First, it breaks the vertical integration by separating the network’s control logic (the control plane) from the underlying routers and switches that forward the traffic (the data plane). Second, with the separation of the control and data planes, network switches become simple forwarding devices and the control logic is implemented in a logically centralized controller, simplifying policy enforcement and network configuration and evolution. A simplified view of this architecture is shown in Fig. 1. It is important to emphasize that a logically centralized programmatic model does not postulate a physically centralized system.
In fact, the need to guarantee adequate levels of performance, scalability, and reliability would preclude such a solution. Instead, production-level SDN network designs resort to physically distributed control planes ,GoalSDN is built dynamically to be able to involve several kinds of network technology, making it ideal for the high-bandwidth and 5g applications. Also designed to separate the network control and forwarding functions. make the network more flexible and agile to support the FunctionalitiesSDN supports multiple functionalities as its centralized controller and separated data and control plane. The SDN’s functionalities, along with its layers and planes, are shown in Figure 2 The general Functionalities of SDN are as follows:Programmability: Network control is directly programmable as the control plane is decoupled from the forwarding or data plane. SDN allows the control plane to be programmed using different software development tools along with the function of customization of the control network according to user requirements.
Centrally Managed: In an SDN, the controller network is logically centralized, thus providing a comprehensive view of the network that appears to the applications or users as a logical device.Flexibility. SDN provides flexibility to network managers. Network managers can manage, configure, secure, and optimize network parameters very rapidly through dynamic, automated SDN programs.
This helps the controllers respond to traffic variations. As controllers run in software, SDN affords the flexibility of synchronization through the network operating system (NOS) approach on different physical or virtual hosts.Granularity. Since networking is spreading across different protocol layers and the level of data flow is aggregating as well, SDN has the features to control the traffic flow with different granularity on the protocol layers and at the aggregate level. These can vary from the core networks to a single connection in a home LAN.Protocol Independence. SDN has a key feature called protocol independence.
It helps run or control a variety of networking protocols and technologies on different SDN network layers. It also enables one to change policies from old to new technologies and supports different protocols for different applications.Open Standard-Based. Instead of multiple vendor devices and protocols, SDN controllers simplify network operation and design based on controller instructions applied through an open standard.
Ability of Dynamic Control. SDN has the ability to modify the network traffic flow dynamically. Dynamic reconfiguration covers wide-area networks, and in data center networks, where constant or continuous transportation of real or virtual machines and their network control schemes need to change in minutes or even seconds.Difference between NFV and SDNNFV and SDN are corresponding approaches, NFV alter services to be virtual oriented but doesn’t provide policies to automate the environment.
SDN is responsible for forwarding packets from one network device to another, while NFV allows routing control functions to run on a virtual machine in a specific server, for example.How SDN and NFV work togetherSDN and NFV are harmonious approaches. Both can offer a new design managing network services. They are enabling a drastic change to take place in network architecture, allowing traditional structures to be broken down into customizable elements that can be chained together programmatically to provide just the right level of connectivity, with each element running on the architecture of its choice. This is the concept of network slicing that will enable networks to be built in a way that maximizes flexibility. Being able to deliver the wide variety of network performance characteristics that future services will demand is one of the primary technical challenges faced by service providers today.
The performance requirements placed on the network will demand connectivity in terms of data rate, latency, QOS, security, availability, and many other parameters — all of which will vary from one service to the next. But future services also present a business challenge: average revenues will differ significantly from one service to the next, and so flexibility in balancing cost-optimized implementations with those that are performance-optimized will be crucial to profitability.Network SlicingNetwork slicing is the ability to deliver multiple virtual network occurrences to be created on top of a common shared infrastructure, meanwhile improving the flexibility and agility.
10 In different words we can define it as an end-to-end logical network supplied by virtual resources runs on a shared physical infrastructure providing service quality. These logical networks can provide and fulfil users with different services depending on the communication requirements. Network slicing provides a network-as-a-service (NaaS) model that supplies 5G diverse communication scenarios by making use of flexibly allocated resources according to the dynamic demands. The idea behind the Network slicing in virtual network is the same as software defined networking (SDN) and network functions virtualization (NFV) in fixed networks.
They were developed and commercially deployed to deliver greater network flexibility by allowing traditional network architectures to be partitioned into virtual elements that can be linked. Virtual Networks can be customized to meet the specific needs of the applications and services. This means that network operators will be able to individually slice the network and customize them depending on the user’s needs, which provides flexibility and elasticity. It results in individually programmable layers within the infrastructure called network slices. Each of these represents an individual network, and they may even have conflicting characteristics. Within the shared network infrastructure, a slice can be used for one industry, for a specific need, or even at a specific time. You can watch a TV series at HD and network slice will provide mobile broadband on the other hand another slice could be connected by with a driverless car which will be connected to a traffic jam and even help with breaking.
How does Slicing occur?Deployment of network slicing enables a single physical network to be sliced into multiple virtual networks that can support different radio access networks (RANs) in the purpose of enhancing device capability, and increase user experience 8The right slice for everyone.A Slice-based 5G enabled some features when comparing to a 4G or a traditional networks, It guarantee smooth process flows, providing better performance than one-size-fits-all networks. Slicing can be scaled up or down as service requirements, As result, everyone will get the right slice in the network. Resources with slicing in the new technology can be isolated of one service to another 9 The architecture of Network slicing in a 5G network is based on an as-a-service basis, so that every service is customized to get best use of necessary requirement of elements. 10 In which each individual slice, has independent characteristics to deliver a specific service type with sharing resources in between. 11 Sub Section Network Slicing in 5GNetwork slicing is expected to be the key component of future 5G networks since the plenty of use cases and new services 5G will support.
These use cases and services will place different requirements on the network in terms of functionality, and their performance requirements will vary enormously. Slice-based 5G has significant advantages than traditional networks:Network slicing can provide logical networks with high throughput due to core architecture when compared to one-size-fits-all paradigm.Network Slice is flexibly changed based on service requirement and number of users demands.Network resources of one service can be isolated from one to another, in view of the fact that each service’s use of resources can vary differentially.
Although different configurations required by services, they cannot affect each other. For that reason, security and reliability intensify.Allocation of physical network resources optimized as a result of network slicing.Fifth-generation networks need to combine multiple services to provide best performance requirements – such as high throughput, low latency, high reliability, high flexibility- into a single physical network infrastructure, and provide each service it’s own customized logical network which is the role of network slicing. The third generation Partnership Project (3GPP) has identified network slicing as the key component to achieve future 5G network goals.
A lot of projects initiated by the 3GPP team can give us a good example of network slicing. for example, the dedicated core (DECOR) feature supports the operator to deploy multiple dedicated core networks by sharing common Public Land Mobile Network (PLMN).A Network slice is a self-contained network has its own resources, topology and provisioning rules.
Various network slices can be sufficient to meet the specific communication needs of different users in future mobile network systems. For example, an driverless car will rely on V2X (vehicle-to-anything) communication which requires low latency but not necessarily a high throughput, on the other hand streaming service is running which require a high bandwidth and is susceptible to latency. Both will be delivered over the same physical network on virtual network slices, to optimise use of the physical network.The new bandwidth-hungry services enabled by 5G, as well as High definition (HD) video, virtual reality (VR) and augmented reality (AR). The high bandwidth requirement of these services are not only challenging the continuing of this technology, but the guarantee of high throughput performance. Moreover, providing high bandwidth with strict packet loss tolerances and high mobility is also difficult for the current networks which designed to support the best performance.
What is a network slice?Taxonomy graph of 5G as evolution of the different tracks of 3G to 4G to 5G:The mobile networks of the last few years have seen a drastic evolution that started with the rise of the internet and the development of smartphones that were capable of of delivering internet content and services to the mobile users. This evolution brought new requirements to the network and leaded to mobile networks that were optimized for packet switched instead of circuit switched based traffic. Especially in the third generation mobile networks (3G) with UMTS and HSxPA the focus was set on the optimization of packet switched traffic.
The further development of mobile networks focused on bringing in an all-IP core architecture and led to the development of the fourth generation of mobile networks (4G) with the Long Term Evolution (LTE) standard being the outcome.Figure 1 Evolution of mobile networks2 Analysis of existing solutionsThe currently used and deployed 4th generation mobile networks, namely those conforming the set of Long Term Evolution (LTE) standards, were designed to meet the requirements of bringing in an all-IP core architecture to serve voice and conventional mobile broadband services. LTE’s architectural design included the reuse and improvement of network functions already used in previous 2nd and 3rd generation mobile networks. Since the most important factors in the development of the LTE core architecture were to increase spectral efficiency and with that the improvement of data rates, the requirements for the architectural design of LTE are just a subset of the requirements of the new 5th generation mobile networks that have to meet the various different requirements of the different services and use cases of future mobile networks.
These use cases are defined by the 3GPP standardization body and can be classified in five categories. An enhanced mobile broadband to meet the demands of the users to stream videos in ultra high definition(UHD), 3D or other contents like virtual reality contents.2.1 Advantages & limitations specialized hw provides good performance but is not flexible:5G will be driven by softwareNetwork functions virtualization (NFV) and software-defined networking (SDN) provide examples for possible new design principles to allow more flexibility and tighter integration with infrastructure layers, although performance and scalability need further investigation. Both approaches stem from the IT realm: NFV leverages recent advances in server virtualization and enterprise IT virtualization; SDN proposes logical centralization of control functions and relies on advances in server scale out and cloud technologies. However, none of those is essentially a networking technology, as the network is assumed to be there, before NFV or SDN can be even used.
Hence, 5G will provide a unified control for multi-tenant networks and services through functional architectures deployment across many operators’ frameworks, giving service providers, and ultimately prosumers, the perception of a convergence across many underlying wireless, optical, network and media technologies. 5G will make possible the fundamental shift in paradigm from the current “service provisioning through controlled ownership of infrastructures” to a “unified control framework through virtualization and programmability of multi-tenant networks and services”.Transformation of SDN 2.2 Key issues specialized hw ? less possibilities to make use of virtualizationFlexibility – all stuff slicing 2.3 Alternatives on how key issues can be solved with the advancements SDN enabled by default through architectureConclusionReference List3GPP TS 23.
0ETSI GS NFV 002 v1.2.15G PPP Architecture Working Group: View on 5G Architecture (Version 2.0)5G Network Architecture and FMC Joe Wilke, Ericsson: Available from: https://www.itu.int/en/ITU-T/Workshops-and-Seminars/201707/Documents/Joe-Wilke-%205G%20Network%20Architecture%20and%20FMC.pdf 1st Dec.17Xin Li et.
al., 2017,„Network Slicing for 5G: Challenges and Opportunities” IEEE Internet Computing, Vol. 21, Issue: 5, pp. 20 – 27, Available from: IEEE Xplore Digital Library 15th Dec. 2017J. Feng, Q.
Zhang, G. Dong, P. Cao and Z. Feng, “An Approach to 5G Wireless Network Virtualization: Architecture and Trial Environment,” 2017 IEEE Wireless Communications and Networking Conference (WCNC), San Francisco, CA, 2017, pp. 1-6.Junseok Kim, Dongmyoung Kim, Sunghyun Choi, 3GPP SA2 architecture and functions for 5G mobile communication system, In ICT Express, Volume 3, Issue 1, 2017, Pages 1-8, ISSN 2405-9595, https://doi.
Sacha Kavanagh,,Network Slicing: https://5g.co.uk/guides/what-is-network-slicing/, “what is network slicing?” January, 2018https://www.telekom.com/en/company/details/network-slicing-485776https://www.cio.
org/NFV/NFV_White_Paper.pdfBenzekki Kamal et al.Software?defined networking (SDN): a survey., Security and Communication Networks 9, no. 18 (2016): 5803-5833.
pdfhttp://ieeexplore.ieee.org/document/7926919/http://www.packetdesign.com/blog/ready-5g-network-slicing/ “Network Functions Virtualisation (NFV); Use NFV is present and SDN is future.
Cases” (PDF). Retrieved 6 June 2014.http://www.3gpp.org/news-events/3gpp-news/1614-sa_5ghttps://books.google.com.eg/books?id=DTsTAAAAQBAJ&lpg=PA13&ots=zS0N2DBnkz&dq=3gpp%20scope&pg=PA8#v=onepage&q=3gpp%20scope&f=falsehttps://eucnc.eu/files/keynotes/Beylat.pdfhttp://onlinelibrary.wiley.com/doi/10.1002/sec.1737/fullhttp://www.5gsummit.org/berlin/docs/slides/HansEinsiedler.pdfhttps://arxiv.org/ftp/arxiv/papers/1704/1704.02129.pdfhttp://ieeexplore.ieee.org/document/7926921/https://www.sciencedirect.com/science/article/pii/S240595951730019Xhttps://arxiv.org/ftp/arxiv/papers/1705/1705.02902.pdfhttps://www.sciencedirect.com/science/article/pii/S2405959515300503http://www.huawei.com/minisite/hwmbbf16/insights/5G-Nework-Architecture-Whitepaper-en.pdfhttps://www.itu.int/en/membership/Documents/missions/GVA-mission-briefing-5G-28Sept2016.pdf https://www.itu.int/dms_pub/itu-r/opb/rep/R-REP-M.2320-2014-PDF-E.pdf3GPP TS23401https://www.sciencedirect.com/science/article/pii/S240595951730019X#br000015https://www.sciencedirect.com/science/article/pii/S2405959515300503https://www.computer.org/csdl/proceedings/wimob/2013/9999/00/06673363.pdfhttp://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7926923 maybe helpshttp://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6515050 power consumption sectionhttp://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7024793 power consumption section